Prognostic value of N-terminal prohormone brain natriuretic peptide for in-hospital and long-term outcomes in patients with infective endocarditis.

Background Limited research studies with a large sample size were performed to evaluate the prognostic value of N-terminal pro-B-type natriuretic peptide (NT-proBNP) for in-hospital or long-term poor outcomes in patients with infective endocarditis. Methods A total of 703 patients with infective endocarditis were enrolled and divided into four groups according to admission NT-pro-BNP (pg/mL) quartiles: Q1 (<258), Q2 (258-1054), Q3 (1055-3522) and Q4 (>3522). Multivariate regression was used to determine independent risk of NT-proBNP for in-hospital and one-year death. Results In-hospital death occurred in 9.0% of patients. The in-hospital mortality was increased from the lowest to the highest NT-proBNP quartiles (1.1%, 3.4%, 9.1% and 22.3%, P < 0.001, respectively). During the one-year follow-up period, 29 patients died: 0 in Q1, 7 (4.6%) in Q2, 8 (5.7%) in Q3 and 14 (12.0%) in Q4 ( P < 0.001). Log-transformed (lg) NT-proBNP had a linear correlation with lg C-reactive protein ( r = 0.308, P < 0.001). Multivariate analysis showed lgNT-proBNP was an independent predictor for both in-hospital (odds ratio 4.59, 95% confidence interval (CI) 2.45-8.61, P < 0.001) and one-year mortality (hazard ratio 3.11, 95% CI 1.65-5.87, P < 0.001). In addition, NT-proBNP had a higher predictive power for in-hospital death than C-reactive protein (area under the curve 0.797 vs. 0.670, P = 0.005). NT-proBNP > 2260 pg/mL had 76.2% sensitivity and 69.1% specificity for predicting in-hospital death. Kaplan-Meier analysis showed that patients with NT-proBNP > 2260 pg/ml had a worse prognosis than those without (log-rank test 18.84, P < 0.001). Conclusion Increased NT-proBNP was independently associated with in-hospital and one-year mortality in patients with infective endocarditis.